JP7646466B2 - Monitoring device, monitoring system, and monitoring method - Google Patents

Description

The present invention relates to a monitoring device, a monitoring system, and a monitoring method.

In refineries, excess gas discharged during the refining process is incinerated in equipment called a flare stack. Flames (flares) and black smoke resulting from the combustion of excess gas are emitted from the tip of the flare discharge section of the flare stack. The flare gas line is connected to piping from multiple devices in the refinery. If the seal of the safety valve of the piping deteriorates, gas that is actually used in the refinery may flow into the flare gas line and be released as a flare from the tip of the flare discharge section. This is undesirable from the viewpoint of energy conservation. When gas that is actually used in the refinery flows in, the size of the flare becomes larger than usual. Taking advantage of this property, the state of the flare is often photographed by a surveillance camera, and a monitor visually checks the image to monitor whether gas that is actually used in the refinery is flowing into the flare gas line.

Patent Document 1 describes a monitoring device that detects abnormalities in flares emitted from a flare stack and the generation of black smoke. The monitoring device described in Patent Document 1 includes an image acquisition unit that acquires images, an image analysis unit that analyzes the images and recognizes a predetermined abnormal flare area set based on the burner unit that emits the flare, and an abnormal flare determination unit that detects abnormal flare based on the size of the flare that occupies the abnormal flare area, calculates the number of pixels in the area where the flare is captured, and the abnormal flare determination unit determines that a flare is abnormal when the number of pixels is equal to or greater than a predetermined threshold value.

JP 2019-175066 A

In a flare stack, it is necessary to grasp the flow rate of flare gas (the increase in the flow rate of flare gas due to the inflow of gas that is originally used in the refinery). The flow rate of flare gas can be monitored by installing a gas flow meter in the flare gas line. However, installing a new gas flow meter in the flare gas line must be carried out during a regular maintenance period for the equipment, and there is a limit to the period during which construction can be carried out. The cost of installing a gas flow meter is high. In addition, in a method in which a flare is photographed with a surveillance camera and a supervisor visually checks it, although it can be estimated that the flare gas flow rate is high based on the size of the flare, it is not possible to quantitatively grasp the flow rate. There is a demand for a simple and low-cost method of measuring the flow rate of flare gas. Patent Document 1 does not describe a method of measuring the flow rate of flare gas.

The present invention aims to provide a monitoring device, a monitoring system, and a monitoring method that can solve the above problems.

According to one aspect of the present invention, the monitoring device includes an image acquisition unit that acquires an image of a flare, an image analysis unit that analyzes the image and identifies pixels in which the flare is captured, a flare volume estimation unit that estimates the volume of the flare based on the pixels in which the flare is captured, a flare gas flow rate estimation unit that estimates the flow rate of the flare gas that generates the flare in a predetermined unit time based on the flare volume calculated for the continuously captured images, and an output unit that compares the flow rate of the flare gas with a predetermined threshold value and issues an alarm.

According to one aspect of the present invention, a monitoring system includes an imaging device that captures video of the area surrounding the flare emission portion of the flare stack, and the above-mentioned monitoring device that acquires the video.

According to one aspect of the present invention, a monitoring method includes the steps of acquiring an image of a flare, analyzing the image and identifying pixels in which the flare is captured, estimating the volume of the flare based on the pixels in which the flare is captured, estimating the flow rate of the flare gas that causes the flare in a predetermined unit time based on the flare volume calculated for the continuously captured images, and comparing the flow rate of the flare gas with a predetermined threshold and issuing an alarm.

According to the present invention, the flow rate of flare gas can be estimated.

FIG. 1 illustrates an example of a monitoring system according to an embodiment. FIG. 2 is a functional block diagram illustrating an example of a monitoring device according to an embodiment. FIG. 2 is a diagram illustrating an example of a video analyzed by a monitoring device according to an embodiment. FIG. 11 is a first diagram illustrating a flare detection process according to an embodiment. FIG. 11 is a second diagram illustrating the flare detection process according to an embodiment. FIG. 11 is a third diagram illustrating the flare detection process according to an embodiment. FIG. 4 is a fourth diagram illustrating a flare detection process according to an embodiment. FIG. 5 is a fifth diagram illustrating a flare detection process in an embodiment. FIG. 11 is a first diagram illustrating a flare volume estimation process according to an embodiment. FIG. 11 is a second diagram illustrating the flare volume estimation process in an embodiment. FIG. 4 is a diagram illustrating a process for estimating a flare gas flow rate in one embodiment. FIG. 4 is a diagram showing an example of actual measured values and estimated values of flare gas flow rates in one embodiment. FIG. 11 is a diagram illustrating an example of a setting process according to an embodiment. FIG. 11 is a diagram showing an example of a process for estimating a flare gas flow rate in one embodiment.

<Embodiment>
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A flare stack monitoring system according to one embodiment will now be described with reference to FIGS.
FIG. 1 is a diagram showing an example of a monitoring system according to an embodiment of the present invention.
Fig. 1 shows the configuration of a surveillance system 1. As shown in Fig. 1, the surveillance system 1 includes a camera 10, a video acquisition device 15, a video distribution device 20, a surveillance monitor 25, a surveillance device 30, and a central surveillance device 40. As shown in the figure, the camera 10 and the video acquisition device 15, the video acquisition device 15 and the video distribution device 20, the video distribution device 20 and the surveillance monitor 25, and the video distribution device 20 and the surveillance device 30 are each connected so as to be able to communicate with each other.

The camera 10 captures images of the flare discharge section at the tip of the flare stack (the tip of the chimney that dissipates the flare. Hereinafter, this will be referred to as the flare discharge section) and its surroundings. The camera 10 outputs the captured images to the image acquisition device 15. The images captured by the camera 10 will be referred to as flare images. The camera 10 is preferably an imaging device that captures moving images, but may also capture still images at a predetermined time interval, for example. The camera 10 may also be an infrared camera. In a flare stack, steam may be injected directly above the flare discharge section 51 to suppress the spread of the flare. In this case, the size of the flare becomes smaller, but the injected steam is also observed instead. Since an infrared camera can detect steam, it is possible to distinguish between the flare and steam and more accurately estimate the flow rate of the flare gas. In addition, an infrared camera can detect hydrogen gas, which is a type of gas that is originally used in oil refineries, when the hydrogen gas is released. The camera 10 may be connected to the image capture device 15 by a wired cable, or may be connected to the image capture device 15 and the monitoring device 30 by wireless communication. In addition, multiple cameras 10 may be provided.

The image capturing device 15 captures the flare image captured by the camera 10 and outputs the flare image to the image distribution device 20.
The video distribution device 20 has a function of distributing video. The video distribution device 20 outputs the flare video acquired from the video acquisition device 15 to the monitoring monitor 25 and the monitoring device 30.
Surveillance monitor 25 displays the flare image. An observer monitors the flare image displayed on surveillance monitor 25. Specifically, the observer looks at the flare image displayed on surveillance monitor 25, checks the surroundings of flare emission section 51 and the sky above, and visually monitors whether the flare is getting larger, whether there is a misfire, whether black smoke is being generated, and the like.

The monitoring device 30 analyzes the flare video captured by the camera 10 and estimates the flow rate of the flare gas. After estimating the flow rate of the flare gas, the monitoring device 30 displays the estimation result and an alarm, etc. on the display unit of the monitoring device 30 or a display device connected to the monitoring device 30, and notifies the monitor. The monitoring device 30 also transmits the flare gas flow rate estimate and the alarm, etc. to the central monitoring device 40. Next, the monitoring device 30 will be described in detail.

FIG. 2 is a functional block diagram illustrating an example of a monitoring device according to an embodiment of the present invention.
The monitoring device 30 is a computer such as a PC (Personal Computer) or a server equipped with a calculation device such as a CPU (Central Processing Unit). As shown in the figure, the monitoring device 30 includes an image acquisition unit 31, a setting information acquisition unit 32, an image analysis unit 33, a flare volume estimation unit 34, a flare gas flow rate estimation unit 35, a storage unit 36, and an output unit 37.

The image acquiring unit 31 acquires the flare image captured by the camera 10. The image acquiring unit 31 outputs the flare image to the image analyzing unit 33.
The setting information acquisition unit 32 acquires various setting information necessary for analyzing the flare image. The setting information acquisition unit 32 records the acquired setting information in the storage unit 36. For example, the setting information acquisition unit 32 accepts an input of information on the width (actual length) of the flare emitting portion 51.
The image analysis unit 33 analyzes the flare image and distinguishes between pixels in which a flare is captured and pixels in which a flare is not captured. Reference is now made to FIG. 3. FIG. 3 is a diagram showing an example of an image analyzed by a monitoring device in one embodiment of the present invention. FIG. 3 shows an example of a flare image 50 captured by the camera 10. Note that the rectangular frames indicated by 52 and 53 do not represent the flare image 50 itself, but represent areas necessary for analyzing the flare image. These areas are arranged based on the setting information acquired by the setting information acquisition unit 32.

(1) For example, the video analysis unit 33 detects the flare emitting portion 51 shown in the flare image by pattern recognition based on its shape. The video analysis unit 33 sets a predetermined position of the detected flare emitting portion 51 (for example, the center of the tip of the flare emitting portion 51) as a reference position in the flare image 50. For example, the video analysis unit 33 records in the storage unit 36 coordinate information of the reference position when the lower left edge of the flare image 50 is set as the origin. Alternatively, instead of the video analysis unit 33 analyzing the image, the monitor may input coordinate information of the tip center of the flare emitting portion 51 to the monitoring device 30.

(2) Based on the setting information acquired by the setting information acquisition unit 32, the video analysis unit 33 sets a predetermined range based on the reference position as the flare monitoring area 52 and the background area 53.
The flare monitoring area 52 defines, for example, the range in which the flare 54 is observed from the tip of the flare emitting portion 51 .
Background area 53 is set in an area where flares and black smoke do not reach. Background area 53 is set, for example, in a part of space that is sufficiently far from flare emitting unit 51 and in which no structures exist. Flare monitoring area 52 and background area 53 can be set arbitrarily by a monitor.

(3) The video analysis unit 33 detects a flare 54 from the flare image 50 based on the RGB luminance values (256 gradation values of red, green, and blue in the RGB color space) corresponding to the flare. The video analysis unit 33 may set the RGB luminance values corresponding to the flare by analyzing the pixels included in the flare monitoring area 52 through processing described below, or a monitor may input the RGB luminance values corresponding to the flare to the monitoring device 30, and the setting information acquisition unit 32 may acquire the values.

(4) The image analysis unit 33 counts the number of pixels in the flare image 50 in which the flare emitting portion 51 appears, the number of pixels corresponding to the width of the flare emitting portion 51 .
(5) The image analysis unit 33 counts the number of pixels in the flare image 50 where the flare 54 appears from a center line passing through the center of the flare emitting portion 51 to the left and right ends of the flare 54 .

The flare volume estimation unit 34 estimates the volume of the flare 54 based on the flare image. More specifically, the flare volume estimation unit 34 assumes that the shape of the right-side flare is the right half of a cylinder for the right-side region when the flare 54 in the image analyzed by the image analysis unit 33 is divided into left and right parts along a vertical center line passing through the center of the flare emission unit 51 from which the flare 54 is emitted, and calculates the volume of each pixel in the up-down direction (height direction) of the flare 54. The flare volume estimation unit 34 also assumes that the shape of the left-side flare is the left half of a cylinder and calculates the volume of each pixel in the up-down direction (height direction) of the flare 54 for the left-side region. The flare volume estimation unit 34 calculates the volumes of the right and left sides of the flare 54 by integrating the volumes for each pixel from the top to the bottom of the flare, and estimates the volume of the entire flare 54 by adding them up.

The flare gas flow rate estimation unit 35 estimates the flare gas flow rate per predetermined unit time based on the total volume of the flare 54 estimated by the flare volume estimation unit 34. More specifically, the flare gas flow rate estimation unit 35 calculates the average of the volume V1 of the flare 54 estimated by the flare volume estimation unit 34 based on the flare image captured at time T1 and the volume V2 of the flare estimated by the flare volume estimation unit 34 based on the image captured at time T2, a certain time ΔT after time T1, estimates this as the instantaneous flare gas flow rate at ΔT, and converts this value into an estimate of the flare gas flow rate in a predetermined unit time TU. The flare gas flow rate estimation unit 35 also compares the estimate of the flare gas flow rate with a preset threshold value, and instructs the output unit 37 to issue an alarm if the flare gas flow rate per unit time or the increase in the flare gas flow rate in successive unit times exceeds the threshold value.

The memory unit 36 stores various information. For example, the memory unit 36 stores coordinate information of the reference position of the flare emission unit 51, setting information of RGB luminance values corresponding to the flare, a calibration curve showing the relationship between the flare gas flow rate and the number of pixels, etc.

The output unit 37 generates an image of a user interface for displaying the flare gas flow rate estimation result by the flare gas flow rate estimation unit 35 and inputting setting information. The output unit 37 also outputs an alarm based on an instruction from the flare gas flow rate estimation unit 35.

4 to 8 are first to fifth diagrams, respectively, illustrating a flare detection process in one embodiment.
FIG. 4(a) shows an example of a part of a flare image 50. FIG. 4(b) shows an image in which only the flare is extracted by analyzing the flare image 50 (hereinafter, referred to as a flare image). The method of creating the flare image shown in FIG. 4(b) is as follows. First, an observer sets the range of RGB luminance values corresponding to the flare as shown in FIG. 5. In FIG. 5, R5 indicates the range of R (red) luminance, G5 indicates the range of G (green) luminance, and B5 indicates the range of B (blue) luminance. Next, the image analysis unit 33 extracts pixels having luminance values included in the set range as pixels indicating a flare by image processing. An image is generated by assigning different colors to pixels indicating a flare and pixels indicating no flare. This results in the flare image in FIG. 4(b). In the flare image in FIG. 4(b), a circular ring-shaped region is displayed as a flare. As can be seen by comparing FIG. 4(a) and FIG. 4(b), only a part of the flare can be detected in the flare image in FIG. 4(b).

In this embodiment, in order to improve the accuracy of flare detection, a three-dimensional space is prepared in which the luminance values of each color of R (red), G (green), and B (blue) of RGB luminance are the coordinate axes. Then, in this three-dimensional space, the RGB luminance values of a representative pixel (flare pixel) in which a flare is reflected are plotted, and the RGB luminance values of a non-flare pixel (background pixel) in which a flare is not reflected are plotted, and the RGB luminance values of each pixel of the flare image 50 are plotted in this three-dimensional space. Here, refer to FIG. 6. Point 61 in FIG. 6 is a point where the RGB luminance values of a representative pixel in which a flare is reflected are plotted. Point 63 is a point where the RGB luminance values of a pixel in which a background such as the sky is reflected are plotted. Point 62 is a point where the RGB luminance values of a pixel in the flare image 50 are plotted. The image analysis unit 33 calculates the distance between the points 62 and 61 (the length of the vector 64; the first distance) based on the coordinates of the point 62 where the RGB luminance values of the pixel in the flare image 50 are plotted and the coordinates of the point 61 where the RGB luminance values of the representative flare are plotted. The image analysis unit 33 also calculates the distance between the points 62 and 63 (the length of the vector 65; the second distance) based on the coordinates of the point 62 where the RGB luminance values of the pixel in the flare image 50 are plotted and the coordinates of the point 63 where the RGB luminance values of the background are plotted. Then, the image analysis unit 33 classifies the pixel corresponding to the point 62 into the one having the shorter length of the vector 64 and the vector 65. That is, if the length of the vector 64 is short, the image analysis unit 33 determines that the pixel corresponding to the point 62 is a pixel indicating a flare (first pixel), and if the length of the vector 65 is short, the image analysis unit 33 determines that the pixel corresponding to the point 62 is a pixel indicating a non-flare (second pixel). The image analysis unit 33 performs the same determination for all pixels in the flare image 50. The image analysis unit 33 creates a flare image by applying different colors to pixels determined to be a flare and pixels determined to be a non-flare. The flare image obtained by this process is shown in FIG. 7(a).

With reference to the flare image in FIG. 7(a), it can be seen that more areas are detected as flares compared to the flare image in FIG. 4(b), and the area of the detected flare has been improved. However, in the flare image in FIG. 7(a), a cavity is generated in the center of the flare. Next, the process of complementing this cavity will be described. First, the video analysis unit 33 divides the flare image shown in FIG. 7(a), for example, from top to bottom in the Y-axis direction into widths each having a predetermined number of pixels, and performs the following process on each of the partial images in a row obtained by dividing the image. The video analysis unit 33 scans the partial images in a row from left to right in the X-axis direction. At that time, if pixels indicating flare are detected consecutively for a predetermined number of pixels or more, then pixels indicating non-flare are detected consecutively for a predetermined number of pixels or more, and then pixels indicating flare are detected consecutively for a predetermined number of pixels or more, the video analysis unit 33 determines that the partial image contains a cavity. Then, the group of pixels not showing flare that are sandwiched between groups of pixels showing flare are recognized as flare, and these groups of pixels are given a color corresponding to the flare (this may be the same color as in FIG. 7(a) or a different color). The flare image obtained as a result of performing this process of filling in the cavities is shown in FIG. 7(b).

Figure 8 (a) shows an example of a flare image 50 different from that shown in Figure 4 (a). In this example, a relatively small flare is emitted from the tip of the flare emitting portion 51. It was confirmed that the method of detecting pixels in which a flare is captured based on the distance between a point showing a representative RGB luminance value of a flare and a point showing an RGB luminance value of the background for each pixel of the flare image 50 described using Figure 6 can detect the flare even when it is small, as shown in the flare image in Figure 8 (b).

After correcting the flare image in this manner, the flare gas flow rate estimation unit 35 calculates the flow rate of the flare gas. When flare gas is supplied to the flare discharge unit 51, the released flare burns out every moment. When flare gas is continuously supplied to the flare discharge unit 51, flares are continuously observed. There is a positive correlation between the flow rate of the flare gas and the size of the flare 54, and it is considered that the flare gas flow rate at the moment the flare image is captured can be expressed as a linear function of the volume of the flare 54 captured in the flare image. Next, a method for the flare volume estimation unit 34 to calculate the volume of the flare from the flare image will be described with reference to FIG. 9.

FIG. 9 is a first diagram illustrating the process of calculating the flare volume in one embodiment.
The setting information acquisition unit 32 acquires the length of the diameter 51L of the flare emitting unit 51 input by the observer and records it in the storage unit 36. The image analysis unit 33 counts the number of pixels corresponding to the diameter 51L based on the pixels in which the flare emitting unit 51 is captured. For convenience of explanation, it is assumed that the flare emitting unit 51 captured in the flare image extends in the vertical direction, and the pixels defining the diameter 51L are arranged in the horizontal direction (X-axis direction) of the flare image. The image analysis unit 33 scans a predetermined position in the Y-axis direction of the flare image (the position in which the flare emitting unit 51 is captured) from left to right in the X-axis direction. The image analysis unit 33 counts the pixels in which the flare emitting unit 51 is captured, and detects the number of pixels of the diameter 51L. For example, the image analysis unit 33 may count the number of pixels in a range indicating the diameter of the flare emitting unit 51 designated by the observer, or may count the number of pixels having a luminance value (luminance value indicating the flare emitting unit 51) designated by the observer. The flare volume estimation unit 34 calculates the size per pixel L_U (the horizontal and vertical lengths of one pixel) by dividing the length of the diameter 51L input by the observer by the number of detected pixels. Next, the flare 54 is divided into the right half and the left half as viewed from the front and the volume is calculated. First, the image analysis unit 33 sets a vertical center line Y (the center line Y is also called the Y axis) that passes through the center of the flare emitting unit 51. The image analysis unit 33 also sets the upper end of the flare emitting unit 51 on the Y axis as the origin O. The image analysis unit 33 counts the number of pixels C_R _y1 from the center line Y at the position y1 in the Y-axis direction of the vertical center line Y to the edge (boundary) of the flare 54 in the horizontal direction to the right. The flare volume estimation unit 34 multiplies the counted number of pixels C_R _y1 by the size per pixel L_U to calculate the flame distance Y_R _y1 . The flare volume estimation unit 34 calculates π×(Y_R _y1 ) ² ×L_U÷2 to calculate the volume of a part of the flare 54 at the position y1 in the Y-axis direction. Here, the above formula is for calculating the right half of the volume of a cylinder when the flame distance Y_R _y1 is the radius and the height is the size of one pixel L_U. For example, the flare volume estimation unit 34 shifts the position y1 by one pixel in the Y-axis direction from the origin O to a position yT corresponding to the top of the flare 54, obtains the volume of the right side of the flare 54 at each position, and sums these. Similarly, the flare volume estimation unit 34 counts the number of pixels C_L _y2 from the center line Y at the position y2 to the edge (boundary) of the flare 54 in the horizontal direction to the left, and multiplies the number of pixels C_L _y2 by the size per pixel L_U to calculate the flame distance Y_L _y2 . The flare volume estimation unit 34 calculates π×(Y_L _y2 ) ² ×L_U÷2 to calculate the volume of a portion of the flare 54 when the height at the position y2 in the Y-axis direction is set to the size of one pixel L_U. The flare volume estimation unit 34 shifts the position y2 by one pixel at a time in the Y-axis direction from the origin O to the position yT corresponding to the top of the flare 54, obtains the volume of the left side of the flare 54 at each position, and sums these. The flare volume estimation unit 34 calculates the volume of the flare 54 by summing the left and right volumes. That is, the flare volume estimation unit 34 calculates the volume V(t) of the flare 54 by the formula (1) in FIG. 9.

In addition, the flare volume estimation unit 34 does not calculate the volume of the flare 54 when the shape of the flare 54 is not symmetrical. In other words, the flare volume estimation unit 34 calculates the ratio of the area of the right side region to the area of the left side region when the flare 54 is divided into left and right by a vertical center line passing through the center of the flare emission unit 51, for example, by dividing the area of the left side region by the area of the right side region, and if the ratio is greater than 1, calculates the inverse of the ratio. If the calculated ratio is equal to or greater than a predetermined threshold, the flare shape is considered to be symmetrical, and estimates the volume of the flare 54. An example of a flare 54 shape that is not symmetrical is shown in FIG. 10. The flare volume estimation unit 34 creates a vector that connects the edge of the flare 54 from the origin O of the vertical center line Y passing through the center of the flare emission unit 51 in the flare image. For example, the flare volume estimation unit 34 creates a vector by shifting the vector direction around the origin by a predetermined angle θ with the center line Y as a reference. The flare volume estimation unit 34 calculates the length of the created vector, and calculates the sum of the lengths of the vectors created in the area on the right side of the center line Y and the sum of the lengths of the vectors created in the area on the left side of the center line Y. In the example of FIG. 10, the flare volume estimation unit 34 creates vectors V1 to V6 and calculates the length of each vector. The flare volume estimation unit 34 calculates the sum L1 of the lengths of the vectors V1 to V2, and calculates the sum L2 of the lengths of the vectors V3 to V6. The flare volume estimation unit 34 calculates, for example, the ratio of L2 to L1, and if the ratio is greater than 1, calculates the reciprocal, and if the calculated ratio is less than a predetermined threshold, determines that the flare 54 is not symmetrical, and if the ratio of L1 to L2 is equal to or greater than a predetermined threshold, determines that the flare 54 is symmetrical. If the flare 54 is not symmetrical, the flare volume estimation unit 34 does not calculate the volume of the flare 54, and if it is symmetrical, calculates the volume of the flare 54 by the process described in FIG. 9.

When the volume V(t) of the flare 54 shown in the flare image captured at time t is calculated based on the flare image, the flare gas flow rate estimation unit 35 estimates the flow rate of the flare gas per a predetermined unit time. Here, FIG. 11 is referred to. The vertical axis of FIG. 11 indicates the volume of the flare 54, and the horizontal axis indicates time. The flare volume estimation unit 34 calculates the volume V of the flare 54 by the above-mentioned method every time a flare image is captured. For example, the flare volume estimation unit 34 calculates the volume V1 from the flare image captured at time T1, and calculates the volume V2 from the flare image captured at the next time T2. When the volume V2 at time T2 is calculated, the flare gas flow rate estimation unit 35 calculates the flow rate N of the flare gas per unit time TU by the following formula (2) using the volumes V1 and V2 calculated based on the flare image captured immediately before.
N=(V1+V2)×ΔT÷2×(TU÷ΔT)・・・(2)

Here, ΔT is the time from T1 to T2. That is, the area of a trapezoid with the volume V1 at time T1 and the volume V2 at time T2 as its top and bottom sides is regarded as the flow rate at ΔT, and this flow rate is converted into the flow rate N per unit time TU. Note that when flare images are continuously acquired, that is, when ΔT is sufficiently short, the time average of the volumes V1 to V2 may be found, and this value may be regarded as the flare gas flow rate, and this value may be converted into the flow rate N per unit time TU (the following formula (2')).
N=(V1+V2)÷2×(TU÷ΔT)...(2')

The flare gas flow rate estimation unit 35 calculates the flow rate of the flare gas per unit time TU each time a new flare image is captured. For example, at time T3, the flare gas flow rate is calculated based on the volume V2 at time T2 and the volume V3 at time T3, and at time T4, the flare gas flow rate is calculated based on the volume V3 at time T3 and the volume V4 at time T4. The same applies to times T5 and after.

Fig. 12 shows the actual measured value of the flare gas flow rate and the flare gas flow rate N calculated by the estimation method described with reference to Fig. 11. The vertical axis of Fig. 12 indicates the flare gas flow rate per unit time TU, and the horizontal axis indicates time. Graph L1 shows the actual measured value of the flare gas flow rate. Graph L2 shows the flare gas flow rate N by the estimation method of this embodiment. As shown in the figure, a graph showing a similar transition to the actual measured value was obtained by the flare gas flow rate estimation method of this embodiment. Therefore, the flare gas flow rate estimation unit 35 calculates an estimated value of the flare gas flow rate by further correcting the flare gas flow rate N, which can be calculated by the above formula (2) or formula (2'), by the following formula (3).
Estimated flare gas flow rate = α × N + β (3)
Here, α and β are any real numbers. α and β are determined by comparison with actual measured values when they are available, knowledge of an expert, engineering knowledge, etc. Then, the flare gas flow rate estimating unit 35 compares the estimated value of the flare gas flow rate calculated by the formula (3) with a predetermined threshold value, and if the estimated value of the flare gas flow rate exceeds the threshold value, instructs the output unit 37 to issue an alarm.

Next, the process of estimating the flare gas flow rate by the monitoring system 1 will be described.
(Setting process)
First, prior to estimating the flare gas flow rate, various setting processes are performed.
FIG. 13 is a diagram illustrating an example of a setting process according to an embodiment.
First, the image acquisition unit 31 acquires the flare image captured by the camera 10 (step S1). The output unit 37 displays the flare images captured by each camera on the monitor of the monitoring device 30. Next, the monitor refers to each flare image and sets the tip position of the flare emitting unit 51 (step S2). For example, the monitor performs input to specify the tip position by inputting coordinate information of the tip. The setting information acquisition unit 32 acquires the input information. When the coordinate information of the tip position is input, the setting information acquisition unit 32 records the coordinate information in the storage unit 36 as setting information of the tip position of the flare emitting unit 51. The setting of the tip position of the flare emitting unit 51 may be performed by the image analysis unit 33 through image analysis.

The monitor then refers to each flare image and sets the flare monitoring area 52 and background area 53 (step S3). The flare monitoring area 52 is set for the purpose of detecting the representative RGB luminance values of pixels in which the flare is captured. The monitor sets a small range near the tip of the flare emitting section 51 as the flare monitoring area 52. This is to ensure that the background other than the flare is not included. The monitor sets the flare monitoring area 52 by inputting the coordinates of the vertices of a rectangular area corresponding to the flare monitoring area 52, for example. The setting information acquisition section 32 records the input coordinate information in the memory section 36 as setting information for the flare monitoring area 52.

The background area 53 is set for the purpose of detecting the RGB luminance values of the background, such as the sky. The observer sets an area in which no flare or smoke is captured as the background area 53. The observer sets the background area 53 by, for example, inputting the coordinates of the vertices of a rectangular area that corresponds to the background area 53. The setting information acquisition unit 32 records the input coordinate information in the memory unit 36 as setting information for the background area 53.

After these settings are made and the monitor performs a specified operation, the image analysis unit 33 calculates the RGB luminance value of each pixel included in the flare monitoring area 52 of the flare image acquired by the image acquisition unit 31. The image analysis unit 33 selects the RGB luminance value of an arbitrary pixel and records this value in the storage unit 36 as the RGB luminance value of a representative flare (corresponding to point 61 in FIG. 6). The image analysis unit 33 also calculates the RGB luminance value of each pixel included in the background area 53 of the flare image acquired by the image acquisition unit 31. The image analysis unit 33 selects the RGB luminance value of an arbitrary pixel and records this value in the storage unit 36 as the RGB luminance value of the background (corresponding to point 63 in FIG. 6). Alternatively, the image analysis unit 33 calculates the average value of the RGB luminance values of each pixel included in the background area 53 and records this value in the storage unit 36 as the RGB luminance value of the background. Alternatively, the RGB luminance values of each pixel included in the flare monitoring area 52 calculated by the video analysis unit 33 may be displayed on the display unit of the monitoring device 30, and the monitor may input appropriate RGB luminance values into the monitoring device 30 while referring to the displayed luminance values, thereby setting the RGB luminance values of a representative flare. The same applies to the RGB luminance values of the background.

Note that a plurality of sets of typical values of the flare's RGB luminance value and the background's RGB luminance value may be prepared in advance according to the time period, season, weather, etc. Then, the observer may select this typical value to set the flare's typical RGB luminance value and the background's RGB luminance value. For example, the storage unit 36 may store typical values such as a typical RGB luminance value of a flare on a sunny day and a typical RGB luminance value of a flare on a rainy day, a typical RGB luminance value of a flare in the daytime and a typical RGB luminance value of a flare in the evening and a typical RGB luminance value of a flare in the background. In addition, when there are a plurality of typical values for the flare's RGB luminance value (for example, an RGB luminance value near the center of the flare and an RGB luminance value near the boundary), a plurality of RGB luminance values may be set.

Next, the monitor sets a threshold value for determining whether or not to issue an alarm in the monitoring device 30 (step S4). The monitor may set a plurality of threshold values according to the flow rate of the flare gas. For example, the monitor sets the threshold value to issue an alarm stating "The flare gas flow rate has exceeded X1" when the flow rate of the flare gas exceeds "X1" Nm ³ /h, and to issue an alarm stating "The flare gas flow rate has exceeded X2" when the flow rate exceeds "X2" Nm ³ /h (X2 > X1). The monitor may also set a threshold value for fluctuations (increases) in the flare gas flow rate. For example, the monitor sets the threshold value to issue an alarm stating "The flare gas flow rate has increased beyond X3" when the relationship between the estimated value N _T1 of the flare gas flow rate at time T1 and the estimated value N _T2 of the flare gas flow rate at time T2, which is the time when the flare gas flow rate is next estimated after time T1, is N _T2 > N _T1 and the difference between them exceeds "X3" Nm ³ /h. The monitor may set a plurality of thresholds according to the increase in the flow rate of the flare gas. The setting information acquisition unit 32 records the input thresholds and the content of the alarm in the storage unit 36.

Then, the observer sets the width of the flare emission section 51 (step S5). For example, the observer inputs the width of the flare emission section 51 in meters. The setting information acquisition section 32 acquires the input width of the flare emission section 51 and records it in the memory section 36.

Then, the monitor sets the width range of the flare emitting section 51 with reference to each flare image (step S6). For example, the monitor performs input to specify the width range of the flare emitting section 51 by inputting coordinate information of the left end and coordinate information of the right end of the flare emitting section 51. The setting information acquisition section 32 acquires the input information. When coordinate information of the left and right ends of the flare emitting section 51 is input, the setting information acquisition section 32 records the coordinate information in the storage section 36 as setting information that specifies the width range of the flare emitting section 51. The width of the flare emitting section 51 may be set by the video analysis section 33 through image analysis.

(Flare gas flow rate estimation process)
Next, the process of estimating the flare gas flow rate will be described. Fig. 14 is a diagram showing an example of the process of estimating the flare gas flow rate in one embodiment. First, the image acquisition unit 31 acquires a flare image captured by the camera 10 (step S11). The image acquisition unit 31 outputs the flare image to the image analysis unit 33. The image analysis unit 33 performs the following process for each pixel of the flare image.

First, the image analysis unit 33 analyzes the RGB luminance values of pixels with a flare image (step S12). Next, the image analysis unit 33 plots the analyzed RGB luminance values in a three-dimensional RGB space with the R luminance value, the G luminance value, and the B luminance value as coordinate axes (step S13). The plotted point corresponds to point 62 in FIG. 6. Next, the image analysis unit 33 calculates the distance from the RGB luminance value of the representative flare and the RGB luminance value of the background (step S14). Specifically, the image analysis unit 33 calculates the distance between the point corresponding to the RGB luminance value of the representative flare (point 61 in FIG. 6) and the plotted point (point 62 in FIG. 6). The image analysis unit 33 also calculates the distance between the point corresponding to the RGB luminance value of the background (point 63 in FIG. 6) and the plotted point (point 62 in FIG. 6). At this time, for example, if representative RGB luminance values of the flare and the background are set for each time period, the video analysis unit 33 reads information on the RGB luminance values of the flare and the background according to the current time period from the storage unit 36 and plots them in three-dimensional space. Also, for example, if multiple representative RGB luminance values of the flare are set, the video analysis unit 33 calculates the distance between the point corresponding to each RGB luminance value and the point plotted in step S13.

Next, the image analysis unit 33 judges whether the point corresponding to the analyzed RGB luminance value is close to a point corresponding to a representative RGB luminance value of the flare (step S15). If it is close to a point corresponding to a representative RGB luminance value of the flare (step S15; Yes), the image analysis unit 33 judges that the pixel analyzed this time is a pixel in which the flare is reflected (step S16). If it is close to a point corresponding to the RGB luminance value of the background (step S15; No), the image analysis unit 33 judges that the pixel analyzed this time is a pixel in which the background is reflected (step S17). Next, the image analysis unit 33 judges whether the analysis has been performed for all pixels of the flare image (step S18), and repeats the process from step S12 until the analysis of all pixels is completed. While repeating the processes of steps S12 to S18, the image analysis unit 33 creates a flare image (FIG. 7(a)) in which pixels determined to be flare and pixels determined to be background are colored differently.

Note that, although the processing of steps S12 to S18 is performed on all pixels contained in the flare image, areas where the flare is clearly not visible, such as areas where flare stack equipment is visible, may be excluded from processing. Alternatively, a monitored area may be set, and steps S12 to S18 may be performed only on pixels within the monitored area.

When the process of determining whether all pixels are flares or background is completed, the image analysis unit 33 next determines whether a cavity exists in the pixel group determined to be a flare (step S19). The image analysis unit 33 divides the flare image vertically into a predetermined number of pixels, and scans each of the divided partial images horizontally. The image analysis unit 33 determines that a cavity exists when the pixel group in which a flare is captured, the pixel group in which a background is captured, and the pixel group in which a flare is captured are detected in this order during scanning. If a cavity exists (step S19; Yes), the image analysis unit 33 performs the cavity completion process described using FIG. 7 on the partial image in which a cavity exists (step S20). Specifically, the image analysis unit 33 updates the flare image by applying a color indicating that the non-flare image group sandwiched between the flare pixel groups is a flare. If a cavity does not exist (step S19; No), the process proceeds to step S21. Next, the video analysis unit 33 determines whether cavity detection processing has been performed for all partial images (step S21), and repeats the processing from step S19 until cavity detection processing and completion processing have been completed for all partial images.

The flare complementation process may involve dividing the flare image horizontally into a predetermined number of pixels, scanning each of the resulting partial images vertically, and determining that a cavity exists in the partial image if pixel groups are detected in the following order: a pixel group that reflects the flare, a pixel group that reflects the background, and a pixel group that reflects the flare.

When the cavity complementation process is completed, the image analysis unit 33 next determines whether the shape of the pixel determined to be a flare is symmetrical between the left and right regions when the shape is divided into left and right regions with respect to the vertical center line (center line Y in FIG. 9) passing through the center of the flare emission unit 51 based on the flare image (step S22). As described with reference to FIG. 10, the flare volume estimation unit 34 creates multiple vectors from the origin O toward the edge of the flare and compares the total length of the vectors created in the left region with the total length of the vectors created in the right region. It calculates the ratio of the total length of the left and right vectors, and if the ratio is greater than 1, it calculates the reciprocal. If the calculated ratio is equal to or greater than a threshold value, the flare volume estimation unit 34 determines that the flare is symmetrical. If the ratio of the total length of the left and right vectors is less than a threshold value, the flare volume estimation unit 34 determines that the flare is not symmetrical. If it is not symmetrical (step S22; No). The process of FIG. 14 ends.

In the case of left-right symmetry (step S22; Yes), the flare volume estimation unit 34 calculates the volume ΔV for each unit height on the left and right sides of the flare 54 (step S23). For example, the flare volume estimation unit 34 calculates the volume of the left half of a cylinder whose radius is the length from the center line Y to the left end of the flare 54 for each pixel in the vertical direction, and calculates the volume of the right half of a cylinder whose radius is the length from the center line Y to the right end of the flare 54. The length from the center line Y to the left end is calculated by the flare volume estimation unit 34 multiplying the number of pixels from the center line Y to the left end by the value obtained by dividing the length input in step S5 of FIG. 13 by the number of pixels in the horizontal direction included in the range of the width of the flare emission unit 51 input in step S6. The same applies to the length from the center line Y to the right end. The flare volume estimation unit 34 sums up the volumes on the left side and the right side to calculate the volume ΔV for each unit height in the vertical direction. Next, the flare volume estimation unit 34 integrates ΔV in the vertical direction of the flare 54 to calculate the volume V (step S24).

Next, the flare gas flow rate estimation unit 35 calculates the time average of the volume V and estimates the flare gas flow rate (step S25). For example, the flare gas flow rate estimation unit 35 calculates the time average of the volume V at ΔT (V1+V2)÷2 based on the volume V1 of the flare 54 calculated based on the flare image taken at time T1 and the volume V2 of the flare 54 calculated based on the flare image taken at time T2, which is ΔT after time T1 (when ΔT is sufficiently short). The flare gas flow rate estimation unit 35 converts the time average of the volume V at ΔT into the total of the volume V per unit time TU, multiplies this by a predetermined coefficient α, and adds a predetermined correction term β to calculate the flow rate of the flare gas.
Flow rate of flare gas=α{(V1+V2)÷2×(TU/ΔT)}+β (3)

Alternatively, when ΔT is long, such as when the interval between images taken by the camera 10 is several minutes, the flare gas flow rate estimation unit 35 estimates the flare gas flow rate by the following formula (3′).
Estimated value of flare gas flow rate = α {(V1 + V2) × ΔT ÷ 2 × (TU / ΔT)} + β
... (3')
It should be noted that α and β are set based on past measured values, engineering knowledge, and the like.

Next, the flare gas flow rate estimation unit 35 compares the flare gas flow rate estimate calculated by formula (3) or formula (3') with the threshold value set in step S4. If the flare gas flow rate estimate exceeds the threshold value, the flare gas flow rate estimation unit 35 instructs the output unit 37 to output an alarm corresponding to the threshold value. Alternatively, if the flare gas flow rate estimate at time T is greater than the flare gas flow rate estimate at time T-ΔT by more than the threshold value, the flare gas flow rate estimation unit 35 assumes that the flare gas flow rate has increased during this period due to deterioration of the seal of the piping, etc., and instructs the output unit 37 to output an alarm. Multiple threshold values for the increase in the flare gas flow rate may be set. The output unit 37 outputs the flare gas flow rate estimate and the alarm (step S26). This makes it possible to estimate the flare gas flow rate based on the flare video captured by the camera 10.

According to the monitoring system 1 of this embodiment, the flow rate of the flare gas can be estimated simply by inputting the flare image into the monitoring device 30. In addition, by setting a threshold value for the increase in the flare gas flow rate, an alarm can be issued in real time when the flare gas flow rate begins to increase. This allows an increase in the flare gas flow rate due to deterioration of the piping or the like to be recognized early and dealt with. In addition, an increase in the flow rate of the flare gas (i.e., the inflow of gas that is originally used in the refinery) can be recognized without unreliable visual confirmation of the flare or the introduction of an expensive flow meter. In addition, as shown in FIG. 1, the flare image can be displayed on the monitoring monitor 25 to perform monitoring as in the past, while referring to the estimated result of the flare gas flow rate notified by the monitoring device 30 and the alarm that was notified. In addition, the monitor can arbitrarily set the flare monitoring area 52, background area 53, threshold value, brightness value corresponding to the flare, etc. according to the properties of the surplus gas and the installation environment of the flare stack, so that it can be applied to any facility.

Each process in the monitoring device 30 can be realized, for example, by the CPU or the like of the monitoring device 30 executing a program. The program executed by the monitoring device 30 may be recorded in a computer-readable recording medium and realized by reading and executing the program recorded in the recording medium. The monitoring device 30 includes hardware such as an OS and peripheral devices. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, an optical magnetic disk, a ROM, a CD-ROM, or a storage device such as a hard disk built into the monitoring device 30. The computer-readable recording medium may also include a medium that dynamically holds a program for a short period of time, such as a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line, or a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system that is the server or client in that case. The above program may also be one for realizing part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

In addition, the components in the above-described embodiment may be replaced with known components as appropriate without departing from the spirit of the present invention. Furthermore, the technical scope of the present invention is not limited to the above-described embodiment, and various modifications may be made without departing from the spirit of the present invention.
Note that an RGB luminance value is a value expressed by a combination of red (R), green (G), and blue (B) luminance values (0 to 255) in the RGB color space, and a black and white luminance value is a value expressed in 256 levels from 0 (black) to 255 (white).

REFERENCE SIGNS LIST 1 Surveillance system 10 Camera 15 Image acquisition device 20 Image distribution device 25 Surveillance monitor 30 Surveillance device 31 Image acquisition section 32 Setting information acquisition section 33 Image analysis section 34 Flare volume estimation section 35 Flare gas flow rate estimation section 36 Memory section 37 Output section 50 Flare image 52 Flare monitoring area 53 Background area

Claims (8)

an image acquisition unit for acquiring an image of a flare;
an image analysis unit that analyzes the image and identifies pixels in which the flare is captured;
a flare volume estimation unit that estimates a volume of the flare based on pixels in which the flare is captured;
a flare gas flow rate estimation unit that estimates a flow rate of the flare gas that generates the flare in a predetermined unit time based on a volume of the flare calculated for the continuously captured images;
an output unit that compares the flow rate of the flare gas with a predetermined threshold value and issues an alarm;
Equipped with
the flare volume estimation unit calculates, for a right-side region when the flare is divided into left and right parts by a vertical center line passing through the center of the emission part of the flare in the image, half the volume of a cylinder whose radius is the distance from the center line to the right end of the flare, for each unit height from the top to the bottom of the flare, and estimates the volume of the right side of the flare by integrating the volumes calculated for each unit height; calculates, for a left-side region, half the volume of a cylinder whose radius is the distance from the center line to the left end of the flare, for each unit height from the top to the bottom of the flare, and estimates the volume of the left side of the flare by integrating the volumes calculated for each unit height; and estimates the volume of the flare by summing the volume of the right side of the flare and the volume of the left side of the flare . the flare volume estimation unit calculates a ratio of an area of a left side region of the flare to an area of a right side region of the flare when the flare is divided into left and right parts by a vertical center line passing through a center of an emission portion of the flare, and calculates the reciprocal of the ratio when the ratio is greater than 1, and estimates the volume of the flare when the calculated ratio is equal to or greater than a predetermined threshold value.
The monitoring device according to claim 1 . The output unit issues an alarm when the flow rate of the flare gas estimated by the flare gas flow rate estimation unit becomes equal to or exceeds a predetermined threshold value or increases by equal to or exceeds a predetermined threshold value.
The monitoring device according to claim 1 or 2 . The image analysis unit classifies each pixel included in the image into either a first pixel close to the flare pixel or a second pixel close to the background pixel based on a luminance value of a predetermined flare pixel in which the flare is captured and a luminance value of a predetermined background pixel in which something other than the flare is captured, and calculates the total number of pixels classified into the first pixel.
The monitoring device according to any one of claims 1 to 3 . the image analysis unit calculates a distance between coordinates represented by RGB luminance values of a detection target pixel, which is one pixel included in the image, and coordinates represented by RGB luminance values corresponding to the predetermined flare, in a three-dimensional space having RGB luminance values as coordinate axes, and determines whether the detection target pixel is classified as the first pixel or the second pixel based on the calculated distance .
5. The monitoring device according to claim 4 . When the detection target pixel classified as the second pixel is located between pixels classified as the first pixel, the image analysis unit classifies the detection target pixel as the first pixel.
The monitoring device according to claim 5 . an imaging device for capturing an image of the vicinity of a flare emission portion in the flare stack;
A monitoring device according to any one of claims 1 to 6 that acquires the video;
A surveillance system comprising: Obtaining an image of the flare;
analyzing the image to identify pixels in which the flare is captured;
estimating a volume of the flare based on pixels in which the flare is captured;
estimating a flow rate of the flare gas causing the flare in a predetermined unit time based on the volume of the flare calculated for the continuously captured images;
A step of comparing the flow rate of the flare gas with a predetermined threshold value and issuing an alarm;
having
the step of estimating the volume of the flare includes calculating, for a right-hand region when the flare is divided into left and right parts by a vertical center line passing through the center of the emission part of the flare in the image, half the volume of a cylinder whose radius is the distance from the center line to the right end of the flare, for each unit height from the top to the bottom of the flare, and integrating the volumes calculated for each unit height to estimate the volume of the right side of the flare; calculating, for a left-hand region, half the volume of a cylinder whose radius is the distance from the center line to the left end of the flare, for each unit height from the top to the bottom of the flare, and integrating the volumes calculated for each unit height to estimate the volume of the left side of the flare; and estimating the volume of the flare by summing the volumes of the right side of the flare and the left side of the flare .

Images